W3C

– DRAFT –
Accessible Platform Architectures Working Group Teleconference

24 May 2023

Attendees

Present
janina, jasonjgw, scott_h, SteveNoble
Regrets
-
Chair
jasonjgw
Scribe
jasonjgw, SteveNoble

Meeting minutes

I'm getting an error message from Zoom claiming the meeting ID is invalid. I'm joining directly from the W3C meeting page.

New Zoom meeting arrangements continued).

Janina: zoom settings for sign language interpreters are in the administrator settings

Interpreter: there are some issues with window spotlighting

Interpreter: when the popout ASL view is used, the interpreter is muted, which is not good

janina: We should use our experience with this new technology to update some of our documents

janina: May take a few weeks to identify all the issues we want to address

janina: Does Raja have any observations?

Raja_Kushalnagar: In some ways the new techniques are an improvement

Raja_Kushalnagar: The speaker detection is good

Raja_Kushalnagar: The interpreter detection is not well implemented yet...it does not work with accuracy

Collaboration Tools Accessibility User Requirements.

SteveNoble: clarification? are we using the new technique or the old?

Roy: Using the old way on this call

scott_h: Some updates have been made

scott_h: Made it more generalized, less screen-reader specific

scott_h: Some technologies have an "accessibility mode" - but there are concerns that accessibility should be baked in, not require someone make a setting change

jasonjgw: Glad to hear we are making progress

jasonjgw: After Janina looks at it, we can discuss how to incorporate it into teh text

jasonjgw: But wondering if implementation-specific issues are in-scope

jasonjgw: We may need to think about that as part of the process

scott_h: Sounds good...we also want to sort out WCAG and ATAG connections

jasonjgw: Bring back next week?

janina: Next week should be good

janina: Concerning overlay technologies...

<janina> A11yEdge Possible Approach here: https://a11yedge.github.io/capabilities/#commandconversion

janina: Would love to have some feedback on this

janina: some details around remapping

scott_h: Sounds good...will read this

jasonjgw: Not sure why proprietary platforms are recasting functionality that is already available in standards-based technologies

jasonjgw: Bring back next week?

scott_h: Yes...sounds good

jasonjgw: Notice that screen-reader bleed-through is occurring in this meeting

jasonjgw: This is new behavior -- sorry about that

Media Accessibility User Requirements.

jasonjgw: Section 2.5 discussion of media objects

jasonjgw: But could use an update

jasonjgw: Something like locating specific navigation points in media was not discussed

jasonjgw: Also, did not notice things like flash mitigation and color re-mapping

janina: Flash mitigation and color contrast we did not mention, because those were already in WCAG

janina: That publication was essentially frozen since 2013

janina: So, lots of updates will be needed

janina: There are also new technologies in development now being discussed by APA

janina: Same is true for EPUB....they are in a position to look at new possibilities such as interlinear translations

janina: Hope to be meeting with a number of groups during TPAC, so we can bring those discussions to bear

janina: The media accessibility requirements work was much easier when we focused on user needs, as opposed to tool requirements

jasonjgw: Some questions on where requirements need to be: authoring, vs. user agent vs. assistive technology requirements

janina: Some issues, like flash mitigation, are life-concerning

jasonjgw: Realize that

janina: Speech recognition: engineers love to emphasize the capabilities of automated captions, but having a script makes that even better

jasonjgw: One revision would be move away from simply HTML requirements to more global on media

Raja_Kushalnagar: It empowers the user to use these automated tools, but there is also a role for the content craeteors

Raja_Kushalnagar: The speaker's intent and the listeners interpretation is not always identical

Raja_Kushalnagar: ASR often does good with text identification, but not the speaker's intent

Raja_Kushalnagar: Matching what the speaker intends and the listener receives needs to be closer

Raja_Kushalnagar: Also, the speaker intent could help inform the generator

janina: There is also the spoken presentation task force which is working on automated speech output

jasonjgw: Not sure if we still have an active github repository to begin opening issues

jasonjgw: Will look into this and email the list

Miscellaneous updates and topics.

Repository: w3c/rqtf

janina: New Tooling is being developed to handle repository tracking

janina: This will help focus on just RQTF documents

ACTION: Roy to test ghurlbot for this channel

<trackbot> Created ACTION-2342 - Test ghurlbot for this channel [on Ruoxi Ran - due 2023-05-31].

<ghurlbot> I created issue #2

<ghurlbot> but I could not add the "action" label.

<ghurlbot> That probably means I don't have push permission on w3c/rqtf.

janina: We will be able to create schemas to help

scott_h: Sounds great!

Summary of action items

  1. Roy to test ghurlbot for this channel
Minutes manually created (not a transcript), formatted by scribe.perl version 210 (Wed Jan 11 19:21:32 2023 UTC).

Diagnostics

No scribenick or scribe found. Guessed: SteveNoble

Maybe present: Interpreter, Raja_Kushalnagar, Roy

All speakers: Interpreter, Janina, jasonjgw, Raja_Kushalnagar, Roy, scott_h, SteveNoble

Active on IRC: janina, jasonjgw, Roy, scott_h, SteveNoble